Influence of back donation effects on the structure of ZnO nanoclusters

Abstract

The structure and properties of ZnO quantum dots is a very popular and rapidly growing field of research for which accurate quantum calculations are challenging to perform. Since the dependence between system size and wall time scales nonlinearly, certain compromises have to be made. A particularly important limiting factor is the size of the basis used, this is especially the case if accurate large calculations are to be carried out. In our work, we discovered that an important O(2p)->Zn(4p) back donation, which greatly influences the strength of the Zn O bond, can be reproduced only if diffuse functions are added to the basis set. We further tested the basis dependence for the magic-sized wurtzite nanophase ZnO clusters which were previously shown to be able to accurately reproduce the magnetically doped II-IV Qdots. In this work, we outline the minimal basis sets required to properly describe Zn O bonds in a nanocluster. It was demonstrated that the rock salt nanophase is incorrectly stabilized if a basis set does not contain sufficiently diffuse functions while the correct wurtzite phase is stabilized when diffuse functions are added. This tendency, similar to that in the ZnO dimer case, was shown to stem from the incorrect lack of Zn(4p) electron density in calculations when using the diffuse-free basis set.

Reply to the ‘Comment on “Realization of the Zn3+ oxidation state”’ by Y. Shang, N. Shu, Z. Zhang, P. Yang and J. Xu, Nanoscale, 2022, 14, DOI: 10.1039/D1NR07031B

“…2− using the VASP code is 5.51 eV (see Table 2), which is in excellent agreement with the experimental value of 5.55 eV. 5 Numerous studies of small clusters have been carried out in the past using the VASP code, which have yielded reliable results in agreement with experiments and with those computed using Gaussian. The interpretation of any properties such as the density of states must be independent of the computational code used.…”

“…The above results as well as the binding energies of Zn in the ZnBeB 22 (CN) 23 and ZnBeB 11 (CN) 12 clusters are found to be nearly the same, namely, 7.04 and 7.09 eV, and 6.33 and 6.37 eV, respectively, for the basis sets 6-311+G(d, p) and 6-311G(d,p). Note that Gutsev et al 5 recently studied the interaction of Zn with O and found that the diffuse functions increase the Zn-O binding energy significantly. The authors argued that this is due to the back donation from Zn 4p to O.…”

Mg-Containing Zn3O3 Structures for Detection of CO2: A DFT Study on CHEM Effects of SERS and Electronic Properties

“…The optimization calculations were also carried out at the B3LYP/6-311+G(d,p) level of theory to explore the effect of a higher basis set on energy and spectral properties. 37 Hereafter, unless mentioned, all further calculations refer to B3LYP/6-31G(d,p) optimized geometries. The optimized geometries of both basis sets were subjected to Hessian calculations, which confirmed that the geometries were stationary points on the respective potential energy surfaces.…”

“…The model substrate structures, Zn 3 O 3 ( P ); Zn 2 MgO 3 ( S ); and Zn 3 O 3 Mg ( I ), and the complex geometries, (1) CZP : (CO 2 interacting with the Zn site of P ); (2) CZS (CO 2 interacting with Zn site of S ); (3) CMS : (CO 2 interacting with Mg site of S ); (4) CZI : (CO 2 interacting with Zn site of I ); and (5) CMI : (CO 2 interacting with Mg site of I ), were optimized using the B3LYP/6-31G­(d,p) , level of computational model (see Tables S1­(a–h)). The optimization calculations were also carried out at the B3LYP/6-311+G­(d,p) level of theory to explore the effect of a higher basis set on energy and spectral properties . Hereafter, unless mentioned, all further calculations refer to B3LYP/6-31G­(d,p) optimized geometries.…”